MIT Professor, MacArthur Genius Fellow, Uses Computer Learning To Predict Cancer Risks

One of the 24 winners of the 2017 MacArthur Genius Award is an MIT Computer Science Professor who uses how machines learn, to teach them to spot signs of cancer earlier.

Regina Barzilay is a leading research in machine learning and natural language processing, she's also now working on applying artificial intelligence to cancer research.

On Sharing The News With A Student

"They allowed me to say it only to one person. I could choose — assuming this person was very discrete. So I selected one person... This is a student who became a dear friend who is currently undergoing treatment for cancer. The reason I felt that I absolutely have to tell her is that I felt that people who are in this condition, they really need to know that there are great things that come after. Even though it doesn't look like it to you now, you would actually feel it later. And I felt that by sharing it, it would give a glimpse of great things that would come for her."

On Natural Language Processing

"So what typically people mean in the field is machines that can perform useful tasks with language. For instance, to translate the document. We can't really say that the machine understands the document in the same way we understand the document. But it can translate it, or it can take a long document and summarize it. Or it can predict the sentiment of a review. So those are very practical applications, or even when you are having a request for your Alexa or any other tool you are using when you are making a request, a machine can translate it into a specific action."

On How Her Cancer Diagnosis Changed Her Work

"In 2014, I was diagnosed with breast cancer. And I was the first one in my family to have any form of cancer. It was just a routine mammogram — 'unexpected' is just an understatement, I couldn't even imagine. But what was even more unexpected was how little data, and machine learning, are used in oncology. I hadn't seen any, and I was treated at an excellent place, at MGH. And I don't think that MGH is an exception, it's actually true from what I understood across the country. Today we are not using at all, data to select treatment, to personalize it, or to help the patient's reduce their uncertainty about the outcomes. I really strongly felt it has to be changed.

Data in medicine is collected for human readers, in the way it is currently collected. And there is very little effort to fully structure and automate it."

Regina Barzilay, MIT Computer Science Professor and MacArthur Genius Award Winner.